Related articles

CMU-MOSEI

Multimodal speech dataset used in the paper's evaluation of audio LLMs and incongruence effects.

• Audio LLMs Make More Emotion Errors When Speech and Text Disagree